Term
| When does an interaction effect occur? |
|
Definition
| An interaction occurs whenever the effect of a predictor variable on a criterion variable is not constant over all of the values of the other independent variables. |
|
|
Term
| In multiple regression, how are interactions represented? |
|
Definition
| interactions are represented by including terms that are products of the predictors (multiplication) |
|
|
Term
In this equation, what is the interaction?
Y = B0 + B1X1 + B2X2 + B3X1X2 + E |
|
Definition
B3X1X2
The Xs are multiplied together to create a new term |
|
|
Term
| When looking at a graphical depiction, how do you know if there is an interaction effect? |
|
Definition
| The lines will not be parallel. |
|
|
Term
| Can you have a significant interaction without significant main effects? |
|
Definition
|
|
Term
| What does mediation mean? |
|
Definition
Variable X has a relation with variable Y.
When we say that variable M is a mediator, we mean that X has an effect on M, and M in turn has an effect on Y.
M is a mediator: An intermediary, a link in a causal chain. |
|
|
Term
| What does moderation mean? |
|
Definition
Variable V is a moderator of the relation of X to Y
We mean that the form or strength of the relation between X and Y depends on the level of V.
To illustrate, there may be a level of V under which the relation of X with Y is much weaker or even absent. |
|
|
Term
| how is mediation typically tested? |
|
Definition
| using multiple regression. p values are examined to see if mediation is present or absent |
|
|
Term
(ab is the indirect effect in mediation)
if ab = .30 and was computed from standardized coefficients, what does this mean? |
|
Definition
| A one standard deviation increase in X yields a .3 standard deviation increase in Y |
|
|
Term
| In mediation, what is the difference between the direct and indirect effects? |
|
Definition
The direct effect is the effect of X on Y.
The indirect effect is the (effect of X on M)(effect of M on Y) |
|
|
Term
| In mediation, what is the total effect? |
|
Definition
| Total effect = indirect effect + direct effect |
|
|
Term
| What are the ways of assigning statistical significance to a mediation effect? |
|
Definition
|
|
Term
| What is the four steps procedure? (and name the steps) |
|
Definition
it's a way of assigning statistical significance to a mediation effect
1) show correlation between x&y
2) show correlation between x&m
3) show correlation between m&y
4) test the effect of x on y controlling for m (should be 0) |
|
|
Term
| the four steps procedure is a way of assigning statistical significance to a mediation effect- name the criticisms |
|
Definition
-Not necessary to establish direct effect first (supression can mask effects).
-We are hypothesizing the null by looking for lack of significance of new direct effect.
-Meaningfulness of just examining p values |
|
|
Term
|
Definition
it's a way of assigning statistical significance to a mediation effect
The Sobel test and variants of it perform a statistical test of the product ab. |
|
|
Term
| What are the criticisms of the Sobel test of mediation significance? |
|
Definition
-The sampling distribution is typically not normal (it is positively skewed)
-This violation makes significance tests less accurate
-positive skew makes them less powerful
-generate high levels of type II errors |
|
|
Term
|
Definition
| it's a way of assigning statistical significance to a mediation effect. Many new samples are created from the data by randomly selecting cases with replacement, ab is estimated from these. |
|
|
Term
| Which test is consistently the most powerful of all mediation tests (Least susceptible to Type II errors)? |
|
Definition
|
|
Term
| What are the criticisms of bootstrapping? |
|
Definition
|
|
Term
| what is Joint Significance Criterion? |
|
Definition
it's a way of assigning statistical significance to a mediation effect.
in order to infer mediation two conditions have to be met: 1) a must be significant, 2) b must be significant |
|
|
Term
| What is a criticisms of the mediation significance test Joint Significance Criterion? |
|
Definition
| JSC does not produce a confidence interval. |
|
|
Term
|
Definition
| the ratio of the indirect effect to the total effect. Effect ratio = ab/ab + c The effect ratio is the proportion of the total effect of X on Y that is mediated by M |
|
|
Term
What would an effect ratio of .65 mean?
(mediation significance testing) |
|
Definition
| 65% of the effect is mediated by M |
|
|
Term
| What are the limitations of effect ratio? |
|
Definition
Descriptive in nature.
Not interpretable if there is suppression (ratio can be below 0 or above 1) |
|
|
Term
| In measurement, what is scaling? |
|
Definition
| represent quantities of attributes numerically |
|
|
Term
| In measurement, what is classification? |
|
Definition
| define whether the objects fall in the same or different categories with respect to a given attribute |
|
|
Term
|
Definition
| classifications made for scientific purposes. They are conceptual abstractions of phenomena that cannot be directly observed |
|
|
Term
| What are tests/scale/measures? |
|
Definition
| An standard instrument to assign meaningful numbers to phenomena |
|
|
Term
| What are the 3 limitations of tests/scales/measures? |
|
Definition
-precision
-accuracy
-utility |
|
|
Term
|
Definition
| To what extent do observations reflect true scores? How much of the scores are due to random error? |
|
|
Term
|
Definition
| How much of the true score is you you think/say it is? |
|
|
Term
| According to classical test theory, how can reliability be represented? |
|
Definition
|
|
Term
| What are the 3 ways of determining reliability? |
|
Definition
1) alternative forms
2) test-retest
3) internal consistency |
|
|
Term
| What is 'alternate forms' and what is a disadvantage? |
|
Definition
| alternate forms is a way of determining reliability, and it is labor intensive |
|
|
Term
| What is 'test-retest' and what is one advantage and one disadvantage? |
|
Definition
way of determining reliability
-easy to use
-not useful for unstable measures |
|
|
Term
| What is 'internal consistency method' and what are the two ways of determining it? |
|
Definition
way of determining reliability
-split half
Cronbach's alpha |
|
|
Term
| What are the 4 factors that affect reliability? |
|
Definition
1) test length
2) item-content
3) sample variability
4) situational characteristics |
|
|
Term
| according to classical test theory, how does reliability affect correlations? |
|
Definition
| less than perfect reliability reduces correlations |
|
|
Term
| what can be done to stop less than perfect reliability reducing correlations? |
|
Definition
| correction for attenuation |
|
|
Term
| What is used to asses validity |
|
Definition
|
|
Term
| What is convergent validity? |
|
Definition
| whether a test correlates with tests of the same construct |
|
|
Term
| What is discriminant validity? |
|
Definition
| whether (or the extent to which) a test correlates with what it should not be measuring |
|
|
Term
| What is criterion validity? |
|
Definition
| whether a test correlates with an external standard (e.g., some critical future behaviour) |
|
|
Term
| what is the term for whether a test correlates with tests of the same construct |
|
Definition
|
|
Term
| what is the term for whether (or the extent to which) a test correlates with what it should not be measuring |
|
Definition
|
|
Term
| what is the term for whether a test correlates with an external standard (e.g., some critical future behaviour) |
|
Definition
|
|
Term
| What does a large T value indicate? |
|
Definition
| a large sample and a large correlation? |
|
|
Term
| What are two-way interactions? |
|
Definition
| Interactions that involve 2 variables. |
|
|
Term
| If you do not have a normal distribution, which method of determining mediation significance should you use? |
|
Definition
|
|
Term
| What is the formula to calculate effect ratio? |
|
Definition
|
|
Term
| What is the formula for the regression equation? |
|
Definition
| Y = B0 + B1(X) + B2(Z) + B3(X*Z) |
|
|
